Abstract:While standard estimation assumes that all datapoints are from probability distribution of the same fixed parameters $\theta$, we will focus on maximum likelihood (ML) adaptive estimation for nonstationary time series: separately estimating parameters $\theta_T$ for each time $T$ based on the earlier values $(x_t)_{t<T}$ using (exponential) moving ML estimator $\theta_T=\arg\max_\theta l_T$ for $l_T=\sum_{t<T} \eta^{T-t} \ln(\rho_\theta (x_t))$ and some $\eta\in(0,1]$. Computational cost of such moving estimator is generally much higher as we need to optimize log-likelihood multiple times, however, in many cases it can be made inexpensive thanks to dependencies. We focus on such example: $\rho(x)\propto \exp(-|(x-\mu)/\sigma|^\kappa/\kappa)$ exponential power distribution (EPD) family, which covers wide range of tail behavior like Gaussian ($\kappa=2$) or Laplace ($\kappa=1$) distribution. It is also convenient for such adaptive estimation of scale parameter $\sigma$ as its standard ML estimation is $\sigma^\kappa$ being average $\|x-\mu\|^\kappa$. By just replacing average with exponential moving average: $(\sigma_{T+1})^\kappa=\eta(\sigma_T)^\kappa +(1-\eta)|x_T-\mu|^\kappa$ we can inexpensively make it adaptive. It is tested on daily log-return series for DJIA companies, leading to essentially better log-likelihoods than standard (static) estimation, with optimal $\kappa$ tails types varying between companies. Presented general alternative estimation philosophy provides tools which might be useful for building better models for analysis of nonstationary time-series.

A Least Squares Estimator for Lévy-driven Moving Averages Based on Discrete Time Observations

Local linear estimator for stochastic differential equations driven by $\alpha$-stable L\'{e}vy motions

The L-1-Norm Kernel Estimator of Conditional Median for Stationary Processes

Least Squares Estimation for the Drift Parameters in the Sub-Fractional Vasicek Processes

Asymptotic Properties of the Least Squares Estimator in Local to Unity Processes with Fractional Gaussian Noise

Least Squares Estimators for Stochastic Differential Equations Driven by Small Lévy Noises

A Least Squares Estimator for Discretely Observed Ornstein–Uhlenbeck Processes Driven by Symmetric Α-Stable Motions

Asymptotic Theory of Least Distances Estimate in Multivariate Linear Models

Local Linear Estimation for Stochastic Processes Driven by $$\alpha $$ Α -Stable L $$\acute{\mathbf{e}}$$ E ´ Vy Motion

Low frequency estimation of continuous-time moving average Lévy processes

Modified least squares estimators for Ornstein–Uhlenbeck processes from low-frequency observations

The Estimation for Lévy Processes in High Frequency Data

Asymptotics of L-1-Estimators in Moving Average Time Series Models

Jump filtering and efficient drift estimation for Lévy-driven SDE's

Local linear estimator for stochastic differential equations driven by $α$-stable Lévy motions

Adaptive exponential power distribution with moving estimator for nonstationary time series

Parameter Estimation for Long-Memory Stochastic Volatility at Discrete Observation

Two-step estimation of ergodic Lévy driven SDE

On the least squares estimation of threshold autoregressive moving-average models

Parameter Estimation of Linear Stochastic Differential Equations with Sparse Observations

ASYMPTOTIC THEORY ON THE LEAST SQUARES ESTIMATION OF THRESHOLD MOVING-AVERAGE MODELS